Thesis. An earnout is not cash at closing. It is a bet, settled later, on numbers the seller no longer controls. Founders accept earnouts to close a valuation gap. What they are usually closing is a confidence gap, and the cost is paid over the years that follow.

01. Month Twenty

The quarterly review ran to schedule, which was one of the things Helen had learned about her new owners: everything ran to schedule. On the screen was a spreadsheet with her product’s name on it, except that it was no longer her product’s name, and the number at the bottom of the column said 71%.

Issue 18 left Helen at the closing table. Her rostering and compliance software for aged-care providers had taken eleven years to build and sixty days to reprice. The buyer’s accountants had rebuilt her recurring revenue and found $1.6 million that did not recur, and rather than cut the headline price, the buyer had moved $4 million of it into an earnout: payable after two years, contingent on the performance of her product inside the buyer’s business, with Helen staying on to deliver it. The deal was announced at the full number. The press release described her as continuing to lead the business through its next phase of growth.

That was month zero.

By month twenty, her product had a new name. It had been folded into the buyer’s care-sector suite and was sold by the buyer’s sales team as one line item among nine. Billing had migrated to the parent’s systems in month seven, and with it went the ability to say, cleanly, which dollars were hers. Two of her senior engineers had been moved to the parent’s platform team, which was behind schedule on a release that had nothing to do with her market. A line had appeared on her profit and loss statement called corporate services allocation. It was not small, and she had not been asked.

Then the renewal price rises arrived, set by the parent’s pricing team and applied across the suite. Two of her longest-standing clients declined to renew. Under the earnout’s client attrition condition, their departure counted against her.

The review was chaired by a divisional president named Marcus, who was courteous, prepared and entirely reasonable. The spreadsheet showed revenue attributable to her product tracking at 71% of the earnout target. Attributable was the word doing the work. Nobody in the room had acted in bad faith. Nobody had needed to.

02. The Bet That Replaced the Price

An earnout is usually explained as a bridge. The seller believes the business is worth one number, the buyer believes it is worth another, and the earnout spans the gap: if the seller’s forecast comes true, she is paid for it. Explained this way, it sounds like optimism formalised. Both sides get to be right.

That is the brochure. The mechanics are less romantic. An earnout does not resolve the disagreement about value. It moves the risk of that disagreement from the buyer to the seller. The buyer pays the undisputed part at closing and pays the disputed part only if the future performs, and the future is now operated by the buyer. The seller has exchanged a certain price, negotiated while she had alternatives, for a contingent one, collectable when she has none.

There are honest uses. A young revenue line a buyer cannot yet underwrite. A genuine difference of conviction about growth, which a seller may be willing to be paid for proving. But the earnout often arrives late in a process, after diligence has found what diligence finds, as a way to preserve a headline that no longer reflects what the buyer intends to pay. It is the mechanism by which everyone keeps the announcement.

The data is blunt about how the bet settles. SRS Acquiom, whose deal terms studies draw on the private-company transactions it administers in the United States, reports that 24% of private-target deals outside life sciences now carry an earnout, up from 19% a decade earlier. And across the deals that carry one, roughly one earnout dollar in five is actually paid. Whatever an earnout is on announcement day, in aggregate it behaves less like price than like risk.

An earnout is a forecast the seller signs and the buyer operates.

03. The Owner Who Became an Employee

Then the ownership changes, and something happens that surprises almost every founder, though it should surprise none. She discovers what it is to work for someone else inside the company she built. Decisions she once made in a corridor now route through committees she does not sit on. The budget is a request. The roadmap is a negotiation. The people she hired report, through one or two intermediate layers, to somebody who did not know the company existed a year ago. The falling out, when it comes, is rarely about a single decision. It is the accumulating discovery that founder is no longer a job the company contains.

I have watched this from the inside. In October 1997, Hummingbird Communications agreed to pay US$60 million for Andyne Computing, the Canadian software company where I was part of the leadership team. The deal made sense on paper, and nobody behaved badly. But the cultures were different, the decision rights moved, and within a couple of years most of the leadership, and much of the team I had worked alongside, had gone. Nearly thirty years later, advising software founders through their exits, the pattern is still recognisable. The acquired team’s tolerance for the acquirer’s way of working is the shortest-dated asset in the deal.

For the founder with an earnout, departure is not only an emotional event. It is a financial one. Many earnout agreements carry leaver provisions: resign before the payout date, or be dismissed for cause, and the contingent money is reduced or extinguished. A founder who leaves in frustration at month fourteen may be walking away from one of the largest components of her own sale price. And the founder who stays past a falling out discovers a quieter problem. Once she is out of the room, she can no longer see the numbers her money depends on. The earnout obligation survives her departure. Her visibility into the business usually does not, unless information and audit rights were written into the sale agreement, and they rarely are, because at signing nobody believed they would be needed.

The earnout survives the founder’s departure. Her visibility into the business does not.

04. When the Buyer Runs the Number

Set bad faith aside. Assume a buyer who wants the acquisition to succeed, which most do. The earnout still sits inside a business the buyer must now run in its own interests, and the buyer’s ordinary decisions move the metric in ways nobody needs to intend.

The product is bundled into a suite, and standalone revenue becomes an allocation performed by the buyer’s finance team. Overhead is charged where overhead has always been charged across the buyer’s divisions, and an earnings target absorbs costs the founder never incurred. Accounting policies are harmonised, as they must be, and the founder’s revenue recognition, built over years of her own judgement calls, is replaced by the parent’s, which defers what she recognised and recognises what she deferred. Readers of Issue 18 will notice the symmetry. Before the sale, Helen’s problem was that she had defined her own number and the buyer rebuilt it. After the sale, the buyer defines the number. The claim she spent a decade making is now adjudicated, quarter by quarter, by the counterparty to the bet.

Engineers are reassigned to the parent’s priorities. Marketing budgets are set by people optimising the whole rather than the part. Prices move to corporate policy, and clients respond to the policy rather than to the founder who no longer sets it. Each decision is defensible. Most are even sensible. The earnout simply delivers their consequences to the seller.

Founders assume the contract protects them here. It protects them exactly as far as its words go, and no further. The most instructive case for a software seller is a decade old and still governs the market. QinetiQ North America bought Cyveillance, a cyber intelligence software company, for US$40 million upfront and up to US$40 million more if revenue reached an agreed level. Revenue did not, and the sellers collected nothing. Their agreement contained the standard protection: the buyer would take no action intended to reduce or limit the earnout. In 2015, the Delaware Supreme Court held that intended meant what it said. It was not enough that the buyer made decisions knowing they would cost the sellers the earnout. The sellers had to prove the buyer acted in order to defeat it, and they could not. The court declined to imply duties the contract did not state. Half the price of the company turned on a single drafting choice: a clause that policed the buyer’s motives instead of prescribing the buyer’s obligations. Express obligations matter. A clause that requires proving the buyer’s state of mind is close to no clause at all.

What the contract does not oblige the buyer to do, the buyer is free not to do.

05. Earning It Once

The defences fall into two groups, and the first group is not about earnouts at all.

The second group is drafting, for the earnout that cannot be avoided.

Measure as high in the profit and loss statement as possible. A revenue target leaves less room for allocation than an earnings target. Define the metric as if the reader will be hostile, because the reader will be the buyer’s finance team.

Ring-fence the measurement: bundled sales, rebranding, cost allocations, transferred staff, discontinued products. Provide for acceleration, so that if the thing being measured ceases to exist, the earnout pays out rather than expires. Negotiate information and audit rights that survive the founder’s employment. Read the leaver clauses as carefully as the price.

Keep the period short, because every added quarter is another season of the buyer’s decisions landing on the seller’s number. And agree the dispute mechanics in advance, expert determination rather than litigation, because the alternative is years in a foreign court against a counterparty with a larger legal budget.

None of this drafting makes an earnout good. It makes it survivable. The order of operations still matters more. The founder who earns the buyer’s confidence before the process needs less of the contract to protect her after it.

06. The Second Sale

Helen sold her company once, at the closing table, at the number everyone announced.

She has been selling it again every quarter since, in a review meeting she does not chair, against a target measured by the other side of the bet.

The first sale took sixty days. The second will take two years.

Only one of them was priced while she could still say no.

THE EARNOUT TEST

Four questions predict most of what an earnout will pay. Is the metric one the seller will still control after the business changes hands? What happens to the target if the buyer renames, rebundles or reprices the product, or moves its people? What does the agreement oblige the buyer to do, not merely refrain from doing, and who can audit the number once the founder has left? And the hardest one: if the earnout paid nothing, would the closing payment alone still be a sale worth signing?

An earnout is where a buyer parks the risk a seller has not retired, and the work that shrinks it happens twelve to twenty-four months before a process begins. Nexxit helps founders find and retire those risks first: the dependence, the contracts, the numbers, the field of buyers. The Assessment is the entry point to Nexxit, Cube Capital’s acquirability programme.

Founders and transactions described in Founder Thesis are composites drawn from recurring patterns in software M&A, except where companies are named.Sources: SRS Acquiom, 2026 M&A Deal Terms Study; Lazard Technology Partners v. QinetiQ North America Operations LLC (Del. 2015).